The present invention relates generally to storage systems, and more specifically, to storage systems that tolerate faults.
In the field of storage systems, a fault is an interruption of service or connectivity. When several disks are serially connected, a fault in one connection may interrupt the service of all disks downstream of the fault. One challenge for storage system design is to include a means of redundancy that increases the fault tolerance of the system without also greatly increasing the cost of the system.